Codesota
.
Papers
Tasks
Benchmarks
Models
OCR
Submit score
Sign in
Submit score
↵
Menu
Codesota · OCR · Benchmarks · robustbench-cifar10-linf
Home
/
OCR
/
Benchmarks
/
robustbench-cifar10-linf
Unknown
robustbench-cifar10-linf
.
OCR benchmark
§ 01 · Robust Accuracy
Robust Accuracy.
Higher is better
#
Model
Score
Source
★
Peng et al. 2023 (WRN-70-16)
Fetched from CodeSOTA API on 2026-04-20
71.07%
codesota-api
2
Wang et al. 2023 (WRN-70-16)
Fetched from CodeSOTA API on 2026-04-20
70.69%
codesota-api
3
Gowal et al. 2021 (WRN-70-16)
Fetched from CodeSOTA API on 2026-04-20
66.11%
codesota-api
§ Related · Explore
More OCR content.
Verified Model Reviews
dots.ocr 3B — 88.41 OmniDocBench, 100+ languages
Mistral OCR 3 — 79.75 composite, verified results
clearOCR — Traditional OCR solution
Rys OCR — Polish SOTA model
Comparisons & Guides
PaddleOCR vs Tesseract comparison
GPT-4o vs PaddleOCR comparison
Docling Tutorial: PDF to Markdown
All OCR Vendors Comparison
View all OCR benchmarks →
←
Back to All Benchmarks